Plant Leaf Disease Detection Using Extended CNN with Parrot Optimizer for Sustainable Farming

Sustainable farming requires early and accurate detection of plant diseases, which is critical for crop health and yield. In this article, we present an extended Convolutional Neural Network (CNN) model combined with the Parrot optimizer for successful detection and classification of plant leaf diseases. The enhanced CNN architecture is designed to capture detailed leaf patterns and characteristics, increasing its detection accuracy across various disease types. The Parrot optimizer, known for its faster convergence and stability, fine-tunes the network's settings, reducing computational complexity and training times. Experimental findings show that our methodology significantly outperforms standard optimization methods in disease categorization, achieving an accuracy of 95.2% after class balancing, compared to 88.5% before. Precision and recall also improved, with values reaching 94.5% and 93.8%, respectively. This model can be a valuable tool for farmers, promoting sustainable agriculture by reducing pesticide use and improving crop management.